Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Lessons after a Half-billion GPT Tokens

kenkantzer.com

Post reply on HN
Print view View original

171–179 of 179 posts

Re: Lessons after a Half-billion GPT Tokens

#171

Post by phillipcarter »

Kiro wrote: post #135
phillipcarter wrote: post #84

Earlier quoted context omitted.

> I wonder why? It seems to work pretty well for me. I read this as "what we do works just fine to not need to use JSON mode". We're in the same boat at my company. Been live for a year now, no need to switch. Our prompt is effective at getting GPT-3.5 to always produce JSON.

There's nothing to switch to. You just enable it. No need to change the prompt or anything else. All it requires is that you mention "JSON" in your prompt, which you obviously already do.

You do need to change the prompt. You need to explicitly tell it to emit JSON, and in my experience, if you want it to follow a format you need to also provide that format.

I've found that this is pretty simple to do when you have a basic schema and there's no need to define one and enable function calling.

But in one of my cases, the schema is quite complicated, and "model doesn't produce JSON" hasn't been a problem for us in production. There's no incentive for us to change what we have that's working very well.

Re: Lessons after a Half-billion GPT Tokens

#172

Post by ProjectArcturis »

wavemode wrote: post #153
visarga wrote: post #140

Earlier quoted context omitted.

> given it’s not reasoning at all When you train a model on data made by humans, then it learns to imitate but is ungrounded. After you train the model with interactivity, it can learn from the consequences of its outputs. This grounding by feedback constitutes a new learning signal that does not simply copy humans, and is a necessary ingredient for pattern matching to become reasoning. Everything we know as humans c…

Yeah but this doesn't change how the model functions, this is just turning reasoning into training data by example. It's not learning how to reason - it's just learning how to pretend to reason, about a gradually wider and wider variety of topics. If any LLM appears to be reasoning, that is evidence not of the intelligence of the model, but rather the lack of creativity of the question.

What's the difference between reasoning and pretending to reason really well?

Re: Lessons after a Half-billion GPT Tokens

#173

Post by fnordpiglet »

ProjectArcturis wrote: post #172
wavemode wrote: post #153

Earlier quoted context omitted.

Yeah but this doesn't change how the model functions, this is just turning reasoning into training data by example. It's not learning how to reason - it's just learning how to pretend to reason, about a gradually wider and wider variety of topics. If any LLM appears to be reasoning, that is evidence not of the intelligence of the model, but rather the lack of creativity of the question.

What's the difference between reasoning and pretending to reason really well?

It’s the process by which you solve a problem. Reasoning requires creating abstract concepts and applying logic against them to arrive at a conclusion.

It’s like saying what’s the difference between between deductive logic and Monte Carlo simulations. Both arrive at answers that can be very similar but the process is not similar at all.

If there is any form of reasoning on display here it’s an abductive style of reasoning which operates in a probabilistic semantic space rather than a logical abstract space.

This is important to bear in mind and explains why hallucinations are very difficult to prevent. There is nothing to put guard rails around in the process because it’s literally computing probabilities of tokens appearing given the tokens seen so far and the space of all tokens trained against. It has nothing to draw upon other than this - and that’s the difference between LLMs and systems with richer abstract concepts and operations.

Re: Lessons after a Half-billion GPT Tokens

#174

Post by peter_d_sherman »

>"Lesson 2: You don’t need langchain. You probably don’t even need anything else OpenAI has released in their API in the last year. Just chat API. That’s it.

Langchain is the perfect example of premature abstraction. We started out thinking we had to use it because the internet said so. Instead, millions of tokens later, and probably 3-4 very diverse LLM features in production, and our openai_service file still has only one, 40-line function in it:

def extract_json(prompt, variable_length_input, number_retries)

The only API we use is chat. We always extract json. We don’t need JSON mode, or function calling, or assistants (though we do all that). Heck, we don’t even use system prompts (maybe we should…). When a gpt-4-turbo was released, we updated one string in the codebase.

This is the beauty of a powerful generalized model – less is more."

Well said!

Re: Lessons after a Half-billion GPT Tokens

#175

Post by Yacovlewis »

mind-blight wrote: post #68
gdiamos wrote: post #21

Earlier quoted context omitted.

rare words are out of vocab errors in vectors Especially if they aren’t in the token vocab

Even worse, named entities vary from organization to organization. We have a client who uses a product called "Time". It's software time management. For that customer's documentation, time should be close to "product" and a bunch of other things that have nothing to do with the normal concept of time. I actually suspect that people would get a lot more bang for their buck fine tuning the embedding models on B2B datas…

Great example of how an entity like that could throw effective RAG out the window

Re: Lessons after a Half-billion GPT Tokens

#176

Post by visarga »

wavemode wrote: post #153
visarga wrote: post #140

Earlier quoted context omitted.

> given it’s not reasoning at all When you train a model on data made by humans, then it learns to imitate but is ungrounded. After you train the model with interactivity, it can learn from the consequences of its outputs. This grounding by feedback constitutes a new learning signal that does not simply copy humans, and is a necessary ingredient for pattern matching to become reasoning. Everything we know as humans c…

Yeah but this doesn't change how the model functions, this is just turning reasoning into training data by example. It's not learning how to reason - it's just learning how to pretend to reason, about a gradually wider and wider variety of topics. If any LLM appears to be reasoning, that is evidence not of the intelligence of the model, but rather the lack of creativity of the question.

Humans are only capable of principled reasoning in domains where they have expertise. We don't actually do full causal reasoning in domains we don't have formal training in. We use all sorts of shortcuts that are similar to what LLMs are doing.

If you consider AlphaTensor or other products in the Alpha family, it shows that feedback can train a model to super-human levels.

Re: Lessons after a Half-billion GPT Tokens

#177

Post by Suppafly »

Xenoamorphous wrote: post #63

> We always extract json. We don’t need JSON mode I wonder why? It seems to work pretty well for me. > Lesson 4: GPT is really bad at producing the null hypothesis Tell me about it! Just yesterday I was testing a prompt around text modification rules that ended with “If none of the rules apply to the text, return the original text without any changes”. Do you know ChatGPT’s response to a text where none of the rules…

If you look at any of the cake decorating fail sites, humans make that sort of mistake all the time.

Re: Lessons after a Half-billion GPT Tokens

#178

Post by eru »

gopher_space wrote: post #131
eru wrote: post #104

Earlier quoted context omitted.

Could you elaborate, please?

Instead of `if X == Y do ...` it's more like `enumerate features of X in such a manner...` and then `explain feature #2 of X in terms that Y would understand` and then maybe `enumerate the manners in which Y might apply X#2 to TASK` and then have it do the smartest number. The most lucid explanation for SQL joins I've seen was in a (regrettably unsaved) exchange where I asked it to compare them to different parts of…

Thanks. That sounds interesting.

Re: Lessons after a Half-billion GPT Tokens

#179

Post by ukuina »

egonschiele wrote: post #114
ukuina wrote: post #113

Earlier quoted context omitted.

Have you tried response_format=json_object? I had better luck with function-calling to get a structured response, but it is more limiting than just getting a JSON body.

I haven't tried response_format, I'll give that a shot. I've had issues with function calling. Sometimes it works, sometimes it just returns random Python code.

Using the openai Python library?
Post reply on HN